Indoor environment monitoring alarm method, device, equipment and medium

Abstract

The invention relates to the technical field of smart home, and discloses an indoor environment monitoring alarm method, device and equipment and a medium. The method comprises the following steps: acquiring environment image data, environment state data, user behavior data and monitoring equipment state data obtained by monitoring an indoor environment; performing spatial feature extraction on the environment image data to obtain spatial feature data; performing time feature extraction on the spatial feature data, the environment state data, the user behavior data and the monitoring equipment state data to obtain time feature data; determining a corresponding risk level according to the spatial feature data and the time feature data; and executing corresponding alarm operation according to the risk level. According to the embodiment of the invention, a good monitoring and security effect can be realized.

Description

Technical Field

[0001] The present application relates to the field of smart home technology, and in particular to an indoor environment monitoring alarm method, device, equipment and medium. Background Art

[0002] With increasing awareness of security, more and more families are installing security devices (such as cameras and infrared sensors) in their homes to monitor and alert people to illegal intrusions and thefts. By installing cameras indoors, they can capture real-time video information and output it to the user's mobile phone or home monitor for real-time display, thus enabling real-time monitoring of the indoor environment.

[0003] However, existing methods of using cameras to monitor indoor environments in real time can only record a specific area indoors. Since it's difficult for users to view surveillance footage in real time, alerts to users when illegal break-ins, thefts, and other violations occur are often ineffective, resulting in a relatively poor deterrent effect against such activities. Furthermore, because cameras are fixed in place and their monitoring areas are relatively fixed, their ability to monitor indoor environments is limited, making it difficult to effectively implement targeted monitoring of the indoor environment. Summary of the Invention

[0004] The purpose of this application is to provide an indoor environment monitoring alarm method, device, equipment and medium, which can monitor designated areas such as indoor environments in a targeted manner, identify designated monitoring information in real time and issue corresponding alarm prompts, thereby achieving better monitoring and security effects.

[0005] The present invention provides an indoor environment monitoring alarm method, comprising: Acquire environmental image data, environmental status data, user behavior data, and monitoring device status data obtained from indoor environmental monitoring; Performing spatial feature extraction on the environmental image data to obtain spatial feature data; Performing time feature extraction on the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data to obtain time feature data; determining a corresponding risk level according to the spatial characteristic data and the temporal characteristic data; Execute corresponding alarm operations according to the risk level.

[0006] In some embodiments, before extracting spatial features from the environmental image data, the method further includes: The environmental image data, the environmental status data, the user behavior data and the monitoring device status data are time-aligned and data-fixed-length intercepted to obtain pre-processed environmental image data, environmental status data, user behavior data and monitoring device status data.

[0007] In some embodiments, extracting spatial features from the environmental image data includes: Performing a layer-by-layer convolution operation on the environmental image data to obtain an environmental feature map; the environmental feature map includes contour features and local position features of several target areas in the indoor environment; Performing dimensionality reduction processing on the environmental feature map to obtain the spatial feature data.

[0008] In some embodiments, extracting temporal features from the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data includes: Encoding the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data to obtain encoded feature data; Performing long-short-term memory time series processing on the encoded feature data to obtain long-short-term memory time series feature data containing feature information related to monitoring environment risks, monitoring equipment risks, and user behavior risks; The long short-term memory time series feature data is decoded to obtain the time feature data.

[0009] In some embodiments, the performing long-short-term memory time series processing on the encoded feature data includes: The characteristic information related to the monitoring environment risk, monitoring equipment risk and user behavior risk in the encoded feature data is reorganized and filtered through the input gate and output gate of the long short-term memory gate, and unnecessary information is discarded through the forget gate to obtain the long short-term memory time series feature data.

[0010] In some embodiments, determining the corresponding risk level based on the spatial feature data and the temporal feature data includes: Extracting characteristic information related to monitoring environment risk, monitoring equipment risk, and user behavior risk from both the spatial characteristic data and the temporal characteristic data to obtain environmental risk characteristic data, equipment risk characteristic data, and user behavior risk characteristic data; Performing a weighted sum operation on the environmental risk characteristic data, the device risk characteristic data, and the user behavior risk characteristic data to obtain a comprehensive risk coefficient; Based on the fuzzy comprehensive evaluation method, the risk level corresponding to the comprehensive risk coefficient is determined.

[0011] In some embodiments, performing corresponding alarm operations according to the risk level includes: According to the mapping relationship between the risk level and the alarm operation, the alarm device is triggered to execute the alarm operation corresponding to the current risk level.

[0012] The present invention also provides an indoor environment monitoring alarm device, comprising: The first module is used to obtain environmental image data, environmental status data, user behavior data and monitoring device status data obtained from indoor environment monitoring; The second module is used to extract spatial features from the environmental image data to obtain spatial feature data; A third module is configured to extract time features from the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data to obtain time feature data; A fourth module is configured to determine a corresponding risk level based on the spatial feature data and the temporal feature data; The fifth module is used to perform corresponding alarm operations according to the risk level.

[0013] An embodiment of the present application further provides an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned indoor environment monitoring alarm method when executing the computer program.

[0014] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned indoor environment monitoring alarm method is implemented.

[0015] The beneficial effects of the present application are as follows: by subjecting environmental image data to spatial feature extraction to extract corresponding spatial feature data, and then using a long short-term memory network to capture the temporal dependency between spatial feature data, environmental status data, user behavior data, and monitoring device status data to obtain temporal feature data, and then based on the weights of each risk dimension of the spatial feature data and temporal feature data, weighted summation is performed to obtain a comprehensive risk value, and the risk is divided into multiple levels according to a preset threshold, the risk level corresponding to the current comprehensive risk value is determined, and the corresponding alarm operation is performed according to the risk level. Since multi-source data monitoring is based on environmental image data, environmental status data, user behavior data, and monitoring device status data, it overcomes the defects of relying on single image data for static monitoring, being unable to associate device status with user behavior, and being prone to missing hidden risks. It can integrate environmental, equipment, and behavior data in real time, accurately identify potential risks, and promptly trigger corresponding alarm measures, identify the correlation between environmental anomalies, abnormal user behavior, and intrusion events caused by equipment failures, improve the comprehensiveness of risk assessment, and achieve better monitoring and security effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is an application environment diagram of the indoor environment monitoring alarm method provided in an embodiment of the present application.

[0017] Figure 2 This is a flow chart of the indoor environment monitoring alarm method provided in an embodiment of the present application.

[0018] Figure 3 It is a structural diagram of the indoor environment monitoring alarm device provided in an embodiment of the present application.

[0019] Figure 4 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0021] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps illustrated may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. Terms such as "first" and "second" in the specification, claims, and drawings are used to distinguish similar items and are not intended to describe a specific sequence or precedence.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0023] The indoor environment monitoring alarm method provided in the embodiments of the present application can be executed by a computer device, which can be a terminal device or a server. Among them, the terminal device includes but is not limited to mobile phones, computers, smart home appliances, vehicle-mounted terminals, aircraft, etc. The server can be an independent physical server, or a server cluster composed of multiple physical servers, or a distributed system, or a cloud server. In addition, the information, data, and signals involved in the embodiments of the present application are all authorized by the relevant objects or fully authorized by all parties, and the collection, use, and processing of the relevant data comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0024] Figure 1 This is an application environment diagram of the indoor environment monitoring alarm method provided by the embodiment of this application. Figure 1 , the indoor environment monitoring alarm method is applied to the indoor environment monitoring alarm method. The indoor environment monitoring alarm method includes a terminal 110 and a server 120. The terminal 110 and the server 120 are connected via a network. The terminal 110 can be a desktop terminal or a mobile terminal. The mobile terminal can be at least one of a mobile phone, a tablet computer, a laptop computer, etc. The server 120 can be implemented as an independent server or a server cluster composed of multiple servers. The terminal 110 is used to send environmental image data, environmental status data, user behavior data and monitoring device status data obtained from indoor environment monitoring to the server 120. The server 120 is used to obtain the environmental image data, environmental status data, user behavior data and monitoring device status data obtained from indoor environment monitoring, perform spatial feature extraction on the environmental image data to obtain spatial feature data, perform temporal feature extraction on the spatial feature data, environmental status data, user behavior data and monitoring device status data to obtain temporal feature data, determine the corresponding risk level according to the spatial feature data and the temporal feature data, and perform the corresponding alarm operation according to the risk level.

[0025] It should be understood that Figure 1The application scenarios shown are merely examples. In actual applications, the indoor environment monitoring alarm method provided by the embodiments of the present application can also be applied to other scenarios. For example, the indoor environment monitoring alarm method described above can be directly applied to terminal 110, which is used to obtain environmental image data, environmental status data, user behavior data, and monitoring device status data obtained from indoor environment monitoring, perform spatial feature extraction on the environmental image data to obtain spatial feature data, perform temporal feature extraction on the spatial feature data, environmental status data, user behavior data, and monitoring device status data to obtain temporal feature data, determine a corresponding risk level based on the spatial feature data and temporal feature data, and perform a corresponding alarm operation based on the risk level.

[0026] To facilitate understanding of the indoor environment monitoring alarm method provided in the embodiment of the present application, the application scenario of the indoor environment monitoring alarm method is exemplarily introduced below, taking the execution subject as the terminal 110 as an example.

[0027] Figure 2 This is a flow chart of the indoor environment monitoring alarm method provided by the embodiment of this application. Figure 2 In some embodiments, the method includes but is not limited to steps S201 to S205.

[0028] Step S201: Acquire environmental image data, environmental status data, user behavior data, and monitoring device status data obtained from indoor environment monitoring.

[0029] Environmental image data refers to video streams or image sequences of indoor environments captured by cameras. Specifically, this can be achieved using RGB cameras or infrared cameras. These data are used to capture the outlines of people and objects within the monitored area, as well as their corresponding positional changes. Environmental state data refers to numerical monitoring data obtained by sensors collecting the environmental state of the indoor environment. Examples include temperature and humidity data, smoke concentration data, and light intensity data. Specifically, these data can be collected using temperature and humidity sensors, smoke concentration sensors, and light intensity sensors. These data are used to capture environmental state changes within the indoor environment. User behavior data refers to numerical monitoring data obtained by sensors collecting user behavior. Examples include user movement trajectories and operational habits. Specifically, infrared sensors can be used to collect user movement trajectories, and device usage and trigger times can be used to collect operational habits. This data is used to characterize the behavior of legitimate users. Monitoring device state data refers to device state data of monitoring devices used to monitor the indoor environment. Examples include current and voltage fluctuations and switch status information. This data can be obtained through interaction with monitoring devices and is used to characterize the device state of the monitoring devices used to monitor the indoor environment.

[0030] As some examples, various monitoring devices can be deployed in an indoor environment, connected to an execution entity, to upload relevant environmental monitoring data (environmental image data, environmental status data, user behavior data, and monitoring device status data) to the execution entity in real time. It should be noted that the trigger information for each type of monitoring device is different. For door and window switch detection sensors, environmental monitoring data is uploaded to the execution entity when the sensor detects the door or window is open. For temperature, smoke, and gas sensors, environmental monitoring data is uploaded to the execution entity when the environmental monitoring data reaches a set threshold.

[0031] Step S202: extracting spatial features from the environmental image data to obtain spatial feature data.

[0032] Spatial feature extraction refers to identifying the structural information of the target area from the image. Specifically, a convolutional neural network can be used to extract feature maps layer by layer and then reduce the dimension through the pooling layer to retain the spatial distribution characteristics of the target area.

[0033] As some examples, spatial feature extraction of environmental image data can be performed by inputting the environmental image data into a pre-trained convolutional neural network model, performing a convolution operation on the environmental image data, rearranging the environmental image data into a two-dimensional feature map, and obtaining spatial information at different levels through convolution kernels of different sizes, thereby predicting the structural information of the target area and obtaining spatial feature data.

[0034] Step S203 , extracting time features from the spatial feature data, environmental status data, user behavior data, and monitoring device status data to obtain time feature data.

[0035] Temporal feature extraction refers to capturing the patterns of multi-source data changing over time. Specifically, long-short-term memory networks can be used to perform temporal modeling on the encoded features to screen dynamic information related to risks.

[0036] As some examples, temporal feature extraction is performed on spatial feature data, environmental status data, user behavior data, and monitoring device status data, which can be performed by performing long-short term memory timing processing on the spatial feature data, environmental status data, user behavior data, and monitoring device status data, so as to extract feature information related to monitoring environment risks, monitoring device risks, and user behavior risks based on the temporal dependency relationship among the spatial feature data, environmental status data, user behavior data, and monitoring device status data, and obtain temporal feature data.

[0037] Step S204: determining the corresponding risk level according to the spatial characteristic data and the temporal characteristic data.

[0038] Risk level determination refers to the calculation of the comprehensive risk coefficient based on spatiotemporal characteristics. Specifically, the level thresholds can be divided by weighted fusion of environmental, equipment and behavioral risk characteristics and combined with fuzzy evaluation methods.

[0039] Step S205: Execute corresponding alarm operations according to the risk level.

[0040] Alarm operation refers to the response measures that match the risk level. For example, low risk triggers a local prompt sound, and high risk activates a remote alarm and links safety equipment to implement a graded disposal mechanism.

[0041] In an embodiment of the present application, the execution subject obtains environmental image data, environmental status data, user behavior data and monitoring device status data obtained from indoor environmental monitoring. The environmental image data is processed by spatial feature extraction to extract the corresponding spatial feature data, and then the temporal dependency between the spatial feature data, environmental status data, user behavior data and monitoring device status data is captured through a long short-term memory network, such as the duration of abnormal operation of the equipment or the behavior of personnel detention, to obtain time feature data. Based on the weights of each risk dimension of the spatial feature data and the time feature data, the weighted sum is taken to obtain a comprehensive risk value, and the risk is divided into multiple levels according to the preset threshold value. The risk level corresponding to the current comprehensive risk value is determined, and the corresponding alarm operation is performed according to the risk level. For example, when a high risk is detected, an audible and visual alarm is automatically triggered and a notification is pushed to the user terminal. For medium and low risks, log recording and equipment self-test are started.

[0042] In some embodiments, before performing spatial feature extraction on environmental image data, it also includes: time alignment and fixed-length data interception of environmental image data, environmental status data, user behavior data and monitoring device status data to obtain preprocessed environmental image data, environmental status data, user behavior data and monitoring device status data.

[0043] Time alignment refers to aligning data from different sources to a common time base. This can be achieved through interpolation or resampling. For example, if there is a timestamp discrepancy between the environmental image data and the environmental status data, linear interpolation can be used to adjust the environmental status data to the same time point as the image data, ensuring temporal consistency across multiple sources.

[0044] Data fixed-length truncation refers to unifying time series data of varying lengths into a fixed length. This can be achieved using sliding window truncation or zero padding. For example, when user behavior data varies in length due to varying trigger frequencies, truncation or padding can be performed by setting a fixed time window to ensure consistent input data dimensions for subsequent feature extraction modules.

[0045] During data acquisition, environmental image data may be captured by a camera at a rate of 30 frames per second, while the temperature and humidity parameters in the environmental status data may be collected by a sensor at a frequency of once per second. Through time alignment, the temperature and humidity data can be interpolated to the timestamp of each image frame, ensuring a corresponding relationship between the environmental image data, environmental status data, user behavior data, and monitoring device status data at the same moment. Furthermore, fixed-length data truncation can be performed by truncation of the time-aligned environmental image data, environmental status data, user behavior data, and monitoring device status data using a fixed-length window of 3 seconds. When the length of a particular data type is insufficient, it is padded to a fixed length with zeros. This preprocessing eliminates misalignment in the temporal dimension of the data while ensuring dimensional consistency in the input to subsequent convolutional networks and time series models. Thus, time alignment solves the time synchronization issue of heterogeneous multi-source data, while fixed-length data truncation eliminates differences in input dimensions, providing a standardized data foundation for subsequent feature extraction.

[0046] In some embodiments, spatial feature extraction is performed on environmental image data, including: performing layer-by-layer convolution operations on the environmental image data to obtain an environmental feature map; the environmental feature map includes contour features and local position features of several target areas in the indoor environment; and performing dimensionality reduction processing on the environmental feature map to obtain spatial feature data.

[0047] Layer-by-layer convolution refers to the process of extracting image features through multiple convolutional layers. This can be achieved using a convolutional neural network with different convolution kernels, such as stacking 3×3 or 5×5 convolution kernels layer by layer. This operation gradually expands the receptive field to capture the contours and spatial distribution of the target area. This operation can extract multi-level local features from the original image. Dimensionality reduction is the process of reducing the amount of feature map data. This can be achieved using maximum pooling or average pooling methods. For example, downsampling the feature map using a 2×2 pooling window reduces computational complexity while retaining key spatial information, providing efficient data input for subsequent time series analysis.

[0048] When extracting spatial features from environmental image data, the environmental image data is input into a convolutional neural network, and spatial features of different scales are extracted layer by layer through multiple convolutional layers. For example, the first convolution layer can identify edge information, the second convolution layer can combine edges to form contours, and the third convolution layer further captures the local position associations of the target area. Subsequently, the pooling layer compresses the feature map, for example, reducing the size of the feature map to one-fourth of the original image, thereby generating spatial feature data containing the contour and position features of the target area. This data retains key spatial information while avoiding redundant calculations. Therefore, through layer-by-layer convolution and dimensionality reduction processing, it is possible to dynamically adapt to the morphological changes of target objects in different monitoring scenarios. For example, when an abnormal object appears in the monitoring area, the contour features can accurately reflect its abnormal shape, and the local position features can identify whether the intrusion area is within the high-risk range.

[0049] In some embodiments, temporal feature extraction is performed on spatial feature data, environmental status data, user behavior data, and monitoring device status data, including: encoding the spatial feature data, environmental status data, user behavior data, and monitoring device status data to obtain encoded feature data; performing long-short-term memory time series processing on the encoded feature data to obtain long-short-term memory time series feature data containing feature information related to monitoring environment risks, monitoring device risks, and user behavior risks; and decoding the long-short-term memory time series feature data to obtain temporal feature data.

[0050] Encoding refers to converting data of different formats into a unified feature vector. This can be achieved using a fully connected neural network. The fully connected layer performs a nonlinear transformation on the input data, allowing different types of data to be mapped to the same feature space, facilitating subsequent time series feature extraction. Long-short-term memory time series processing refers to modeling time series data using a long-short-term memory network. This can be achieved through an input gate, output gate, and forget gate structure. The input gate controls the storage of new information, the output gate controls the output of information, and the forget gate controls the forgetting of historical information, thereby screening out key time series features related to risk. Decoding refers to converting the hidden state output by the long-short-term memory network into usable time feature data. This can be achieved using deconvolution or a fully connected layer. The spatial structure of the data is restored through inverse mapping and the final time features are extracted.

[0051] When extracting temporal features from spatial feature data, environmental state data, user behavior data, and monitoring device status data, the data is first fed into the encoding module. A fully connected layer converts the multi-dimensional, heterogeneous data into a uniformly dimensional encoded feature vector, known as the encoded feature data. The encoded feature data is then fed into a long-short-term memory (LSTM) network (LSTM), where a gating mechanism dynamically adjusts the temporal features. For example, when a sudden increase in smoke concentration is detected, the input gate increases the weight of the environmental state data at that moment, while the forget gate reduces the memory strength of the preceding normal data. Finally, the decoding module reconstructs the temporal features into temporal feature data that matches the original data dimensions. For example, the hidden layer output is converted into a feature map containing a risk probability distribution. Thus, the unique gating mechanism of the LSTM network can simultaneously process multi-dimensional temporal data from environmental spatial features, device status, and user behavior. For example, when detecting the simultaneous occurrence of camera occlusion and abnormal movement, the temporal correlation between the two can be accurately identified, avoiding misjudgments based on a single dimension.

[0052] In some embodiments, long-short-term memory time series processing is performed on the encoded feature data, including: reorganizing and filtering the feature information related to monitoring environment risks, monitoring equipment risks and user behavior risks in the encoded feature data through the input gate and output gate of the long-short-term memory gate, and discarding unnecessary information through the forget gate to obtain long-short-term memory time series feature data.

[0053] The input gate is a gating unit used to control the amount of feature information input at the current moment. Specifically, it can be implemented using a combination of the sigmoid and tanh functions. By calculating the importance weights of the input features, key risk-related features can be screened. The output gate is a gating unit used to control the amount of feature information output. Specifically, it can use an activation function to perform a nonlinear transformation on the reorganized features to retain risk characteristics associated with the time series. The forget gate is a gating unit used to remove redundant features. Specifically, it can use a forgetting factor to dynamically attenuate historical states to eliminate interference information irrelevant to the current risk assessment.

[0054] During long- and short-term memory (LSTM) time series processing of the encoded feature data, the encoded feature data is selectively received by the input gate, with higher weights assigned to features relevant to pre-set scenarios. For example, features associated with sudden changes in smoke concentration, abnormal device current fluctuations, and unusual user movement patterns are associated with these features. These selected features are superimposed with historical states in the memory unit, and the output gate maps these superimposed features into risk association patterns over time. Simultaneously, the forget gate continuously monitors non-critical features such as device operating noise and ambient light changes. When such features fail to reach a threshold for multiple consecutive time steps, the corresponding memory states are gradually cleared. This process, through the synergistic effect of the gating units, enables the time series feature data to dynamically reflect the evolution of environmental risks. This triple gating mechanism enables dynamic feature screening. For example, when a monitored device experiences a sudden fault, the input gate quickly captures the abnormal current signature, while the forget gate simultaneously clears historical data from the device's normal state. This allows for a complete temporal evolution of the fault signature, thereby improving the timeliness of device risk identification.

[0055] In some embodiments, the corresponding risk level is determined based on the spatial feature data and the temporal feature data, including: extracting feature information related to the monitoring environment risk, monitoring equipment risk and user behavior risk from both the spatial feature data and the temporal feature data to obtain environmental risk feature data, equipment risk feature data and user behavior risk feature data; performing a weighted sum operation on the environmental risk feature data, the equipment risk feature data and the user behavior risk feature data to obtain a comprehensive risk coefficient; and determining the risk level corresponding to the comprehensive risk coefficient based on a fuzzy comprehensive evaluation method.

[0056] Environmental risk feature data refers to feature information related to indoor environmental safety hazards obtained through image processing and sensor data analysis. Specifically, it can be obtained by using a convolutional neural network to extract the contours of abnormal objects or smoke diffusion features in environmental image data, or by using a long short-term memory network to extract the temporal features between environmental image data, environmental status data, user behavior data, and monitoring equipment status data. It is used to identify environmental risks such as fire, water leakage, or gas leakage. Equipment risk feature data refers to feature information related to equipment failure or abnormal operation obtained by analyzing monitoring equipment status parameters. Specifically, it can be achieved by using sensor data to monitor the temperature, voltage, or signal strength of the equipment. It is used to identify equipment risks such as camera failure, sensor failure, or network interruption. User behavior risk feature data refers to feature information related to abnormal behavior obtained by analyzing user activity data. Specifically, it can be achieved by using a behavior recognition algorithm to analyze the movement trajectory or action pattern of people in the monitoring video. It is used to identify behavioral risks such as illegal intrusion, theft, or dangerous operation.

[0057] To determine the risk level, spatial feature data is first used to extract the distribution characteristics of abnormal objects in the environmental image. For example, a convolutional layer is used to identify the shape of flames or the extent of smoke spread. Temporal feature data is then used to extract the changing trends of sensor parameters, such as the duration of a sudden temperature rise or device signal interruption. User behavior data is then analyzed through time series analysis to capture unusual movement patterns, such as prolonged lingering or rapid climbing. Subsequently, the three risk features are weighted according to preset weights—for example, a weight of 0.5 for environmental risk, 0.3 for device risk, and 0.2 for behavioral risk—to produce a comprehensive risk coefficient ranging from 0 to 1. Finally, a fuzzy set membership function is used to map the comprehensive coefficient to a preset risk level. For example, a coefficient exceeding 0.7 indicates high risk, triggering an audible and visual alarm. This integration of multidimensional features of the environment, devices, and behavior, combined with weighted calculation and fuzzy evaluation, enables the distinction between true risk and interference factors. For example, if a camera detects movement but the device status indicates that the door and window sensors are normal and there are no abnormal temperature changes, the probability of false positives can be reduced. Furthermore, the fuzzy evaluation method avoids the limitations of traditional fixed thresholds and can handle the uncertainty caused by sensor data fluctuations.

[0058] In some embodiments, executing a corresponding alarm operation according to the risk level includes: triggering an alarm device to execute the alarm operation corresponding to the current risk level according to a mapping relationship between the risk level and the alarm operation.

[0059] Once the risk level is determined, the system will query a pre-stored mapping table to determine the corresponding alarm method for that level. For example, if the risk level is level one (e.g., a user waking up at night for a normal nighttime awakening), the app will push an alert message and log it. If the risk level is level two (e.g., smoke detected but no fire confirmed), a local defense strategy will be activated (disabling dangerous equipment, sound and light alarms). If the risk level is level three (e.g., a confirmed fire or intrusion), the system will contact the 110 / firefighting platform and remotely lock control of the house (user authorization required). This process dynamically adjusts the alarm response intensity by matching the risk status with the preset strategy in real time, preventing users from missing high-risk events due to a single alarm method. Thus, by establishing a hierarchical mapping mechanism, alarm operations are precisely aligned with risk levels, avoiding interference caused by excessive alarms for low-risk events while ensuring that high-risk events trigger multiple alarm methods, significantly improving the efficiency of handling abnormal events.

[0060] See also Figure 3 The embodiment of the present application further provides an indoor environment monitoring alarm device, which can implement the above-mentioned indoor environment monitoring alarm method, and the device includes: The first module 301 is used to obtain environmental image data, environmental status data, user behavior data and monitoring device status data obtained by monitoring the indoor environment; The second module 302 is used to extract spatial features from the environmental image data to obtain spatial feature data; The third module 303 is used to extract time features from the spatial feature data, environmental status data, user behavior data, and monitoring device status data to obtain time feature data; The fourth module 304 is used to determine the corresponding risk level based on the spatial feature data and the temporal feature data; The fifth module 305 is used to perform corresponding alarm operations according to the risk level.

[0061] The specific implementation of the indoor environment monitoring alarm device is basically the same as the specific embodiment of the indoor environment monitoring alarm method described above, and will not be repeated here.

[0062] Figure 4 It is a block diagram of an electronic device according to an exemplary embodiment.

[0063] Refer to the following Figure 4 4 to describe the electronic device 400 according to this embodiment of the present disclosure. Figure 4 The electronic device 400 shown is merely an example and should not limit the functionality and scope of use of the embodiments of the present disclosure.

[0064] like Figure 4 As shown, electronic device 400 is implemented as a general-purpose computing device. Components of electronic device 400 may include, but are not limited to, at least one processing unit 410, at least one storage unit 420, a bus 430 connecting various system components (including storage unit 420 and processing unit 410), a display unit 440, and the like.

[0065] The storage unit stores program codes, which can be executed by the processing unit 410, so that the processing unit 410 executes the steps according to various exemplary embodiments of the present disclosure described in the above indoor environment monitoring alarm method section of this specification.

[0066] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 4201 and / or a cache memory unit 4202 , and may further include a read-only memory unit (ROM) 4203 .

[0067] The storage unit 420 may also include a program / utility 4204 having a set (at least one) of program modules 4205, such program modules 4205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0068] Bus 430 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0069] The electronic device 400 may also communicate with one or more external devices 400′ (e.g., a keyboard, a pointing device, a Bluetooth device, etc.), one or more devices that enable a user to interact with the electronic device 400, and / or any device that enables the electronic device 400 to communicate with one or more other computing devices (e.g., a router, a modem, etc.). Such communication may occur via an input / output (I / O) interface 450. Furthermore, the electronic device 400 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 460. The network adapter 460 may communicate with other modules of the electronic device 400 via the bus 430. It should be understood that, although not shown in the figures, other hardware and / or software modules may be used in conjunction with the electronic device 400, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0070] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned indoor environment monitoring alarm method is implemented.

[0071] The indoor environment monitoring alarm method, device, equipment and medium provided in the embodiments of the present application extract the corresponding spatial feature data by subjecting the environmental image data to spatial feature extraction processing, and then capture the temporal dependency between the spatial feature data, environmental status data, user behavior data and monitoring device status data through a long short-term memory network to obtain time feature data. Then, based on the weights of each risk dimension of the spatial feature data and the time feature data, a weighted sum is performed to obtain a comprehensive risk value. The risk is divided into multiple levels according to a preset threshold, the risk level corresponding to the current comprehensive risk value is determined, and the corresponding alarm operation is performed according to the risk level. Since it is based on multi-source data monitoring of environmental image data, environmental status data, user behavior data and monitoring device status data, it overcomes the defects of relying on single image data for static monitoring, which cannot associate device status with user behavior and is prone to missing hidden risks. It can integrate environmental, equipment and behavior data in real time, accurately identify potential risks and trigger corresponding alarm measures in a timely manner, identify the correlation between environmental anomalies caused by equipment failures, abnormal user behavior and intrusion events, improve the comprehensiveness of risk assessment, and achieve better monitoring and security effects.

[0072] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above-mentioned method according to the embodiments of the present disclosure.

[0073] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0074] Computer-readable storage media may include a data signal propagated in baseband or as part of a carrier wave, which carries readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0075] Those skilled in the art will appreciate that the modules described above can be distributed in the device according to the description of the embodiment, or can be modified accordingly to be used in one or more devices that are different from the embodiment. The modules of the above embodiment can be combined into one module or further divided into multiple submodules.

[0076] While the exemplary embodiments of the present disclosure have been specifically illustrated and described above, it should be understood that the present disclosure is not limited to the detailed structures, configurations, or implementations described herein; rather, the present disclosure is intended to encompass various modifications and equivalent configurations within the spirit and scope of the appended claims.

Claims

1. A method for monitoring and alarming an indoor environment, characterized in that: include: Acquire environmental image data, environmental status data, user behavior data, and monitoring device status data obtained from indoor environmental monitoring; Performing spatial feature extraction on the environmental image data to obtain spatial feature data; Performing time feature extraction on the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data to obtain time feature data; determining a corresponding risk level according to the spatial characteristic data and the temporal characteristic data; Execute corresponding alarm operations according to the risk level.

2. The indoor environment monitoring alarm method according to claim 1, characterized in that: Before performing spatial feature extraction on the environmental image data, the method further includes: The environmental image data, the environmental status data, the user behavior data and the monitoring device status data are time-aligned and data-fixed-length intercepted to obtain pre-processed environmental image data, environmental status data, user behavior data and monitoring device status data.

3. The indoor environment monitoring alarm method according to claim 1, characterized in that: The extracting spatial features from the environmental image data includes: Performing a layer-by-layer convolution operation on the environmental image data to obtain an environmental feature map; the environmental feature map includes contour features and local position features of several target areas in the indoor environment; Performing dimensionality reduction processing on the environmental feature map to obtain the spatial feature data.

4. The indoor environment monitoring alarm method according to claim 1, characterized in that: The extracting of time features from the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data includes: Encoding the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data to obtain encoded feature data; Performing long-short-term memory time series processing on the encoded feature data to obtain long-short-term memory time series feature data containing feature information related to monitoring environment risks, monitoring equipment risks, and user behavior risks; The long short-term memory time series feature data is decoded to obtain the time feature data.

5. The indoor environment monitoring alarm method according to claim 4, characterized in that: The performing long-short-term memory time series processing on the encoded feature data includes: The characteristic information related to the monitoring environment risk, monitoring equipment risk and user behavior risk in the encoded feature data is reorganized and filtered through the input gate and output gate of the long short-term memory gate, and unnecessary information is discarded through the forget gate to obtain the long short-term memory time series feature data.

6. The indoor environment monitoring alarm method according to claim 1, characterized in that: The determining of the corresponding risk level according to the spatial characteristic data and the temporal characteristic data includes: Extracting characteristic information related to monitoring environment risk, monitoring equipment risk, and user behavior risk from both the spatial characteristic data and the temporal characteristic data to obtain environmental risk characteristic data, equipment risk characteristic data, and user behavior risk characteristic data; Performing a weighted sum operation on the environmental risk characteristic data, the device risk characteristic data, and the user behavior risk characteristic data to obtain a comprehensive risk coefficient; Based on the fuzzy comprehensive evaluation method, the risk level corresponding to the comprehensive risk coefficient is determined.

7. The indoor environment monitoring alarm method according to claim 1, characterized in that: The performing of corresponding alarm operations according to the risk level includes: According to the mapping relationship between the risk level and the alarm operation, the alarm device is triggered to execute the alarm operation corresponding to the current risk level.

8. An indoor environment monitoring alarm device, characterized in that: include: The first module is used to obtain environmental image data, environmental status data, user behavior data and monitoring device status data obtained from indoor environment monitoring; The second module is used to extract spatial features from the environmental image data to obtain spatial feature data; A third module is configured to extract time features from the spatial feature data, the environmental status data, the user behavior data, and the monitoring device status data to obtain time feature data; A fourth module is configured to determine a corresponding risk level based on the spatial feature data and the temporal feature data; The fifth module is used to perform corresponding alarm operations according to the risk level.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the indoor environment monitoring and alarm method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the indoor environment monitoring and alarm method according to any one of claims 1 to 7 is implemented.

Images